1AMANDA.CONF(5) File formats and conventions AMANDA.CONF(5)
2
6 amanda.conf - Main configuration file for Amanda, the Advanced Maryland
7 Automatic Network Disk Archiver
8
10 amanda.conf(5) is the main configuration file for Amanda. This manpage
11 lists the relevant sections and parameters of this file for quick
12 reference.
13 The file <CONFIG_DIR>/amanda.conf is loaded if it exists then the files
15 <CONFIG_DIR>/<config>/amanda.conf is loaded.
16
18 There are a number of configuration parameters that control the
19 behavior of the Amanda programs. All have default values, so you need
20 not specify the parameter in amanda.conf if the default is suitable.
21 COMMENTS
23 Lines starting with # are ignored, as are blank lines. Comments may be
24 placed on a line with a directive by starting the comment with a #. The
25 remainder of the line is ignored.
26 KEYWORDS AND IDENTIFIERS
28 Keywords are case insensitive, i.e. mailto and MailTo are treated the
29 same. Also, the characters '-' and '_' are interchangeable in all
30 predefined Amanda keywords: device_property and device-property have
31 the same meaning. This manpage uses the dashed versions, but the
32 underscored versions will be accepted for backward compatibility
33 Identifiers are names which are defined in the configuration itself,
35 such as dumptypes or interfaces. Identifiers are are case-insensitive,
36 but sensitive to '-' vs. '_'. Identifiers should be quoted in the
37 configuration file, although For historical reasons, the quotes are
38 optional.
39 Strings are always quoted with double quotes ("), and any double quotes
41 or backslashes within the string are escaped with a backslash:
42 tapelist "/path/to/tapelist"
44 property "escaped-string" "escaping: \\ (backslash) and \" (double-quote)"
45 To summarize, then:
47 # QUOTES CASE -/_
49 logdir "logs" # required sensitive sensitive
50 send-amreport-on strange # prohibited insensitive insensitive
51 tapetype "EXABYTE" # optional insensitive sensitive
52 define dumptype "dt" { # optional insensitive sensitive
54 "dumptype-common" # optional insensitive sensitive
55 strategy noinc # prohibited insensitive insensitive
56 }
57 VALUE SUFFIXES
59 Integer arguments may have one of the following (case insensitive)
60 suffixes, some of which have a multiplier effect:
61 b byte bytes
63 Some number of bytes.
64 bps
66 Some number of bytes per second.
67 k kb kbyte kbytes kilobyte kilobytes
69 Some number of kilobytes (bytes*1024).
70 kps kbps
72 Some number of kilobytes per second (bytes*1024).
73 It is the default multiplier for all size options.
75 m mb meg mbyte mbytes megabyte megabytes
77 Some number of megabytes (bytes*1024*1024).
78 mps mbps
80 Some number of megabytes per second (bytes*1024*1024).
81 g gb gbyte gbytes gigabyte gigabytes
83 Some number of gigabytes (bytes*1024*1024*1024).
84 t tb tbyte tbytes terabyte terabytes
86 Some number of terabytes (bytes*1024*1024*1024*1024).
87 tape tapes
89 Some number of tapes.
90 day days
92 Some number of days.
93 week weeks
95 Some number of weeks (days*7).
96 Note
98 The value inf may be used in most places where an integer is
99 expected to mean an infinite amount.
100 Boolean arguments may have any of the values 1, y, yes, t, true
102 or on to indicate a true state, or 0, n, no, f, false or off to
103 indicate a false state. If no argument is given, true is
104 assumed.
105 PARAMETER ORDER
107 In general, the order in which parameters occur in the configuration
108 file does not matter, with the exception of subsection inheritance. For
109 example, if dumptype "normal-encrypt" which inherits from dumptype
110 "normal", then "normal" must appear first in the configuration file.
111 STRINGS
113 Quoted strings in Amanda follow a common, C-like syntax. Printable
114 characters and whitespace are kept as-is, except that the backslash
115 character (\) is used as an escape character, and a double-quote ends
116 the string. The allowed escape sequences are
117 ESCAPE SEQUENCE BECOMES
119 \\ \
120 \" "
121 \n (newline)
122 \t (tab)
123 \r (carriage return)
124 \f (form-feed)
125 \1 - \7
126 \01 - \77
127 \001 - \377 (character specified in octal)
128 Illegally quoted strings are handled on a "best-effort" basis, which
129 may lead to unexpected results.
130 Examples:
132 finserver "/data/finance/XYZ Corp's \"real\" finances" finance-high eth0 -1
134 property "syspath" "C:\\WINDOWS\\SYSTEM"
135 SUBSECTIONS AND INHERITANCE
137 Amanda configuration files may include various subsections, each
138 defining a set of configuration directives. Each type of subsection is
139 described below. Note that all types of subsections can inherit from
140 other subsections of the same type by naming the "parent" section in
141 the "child" subsection. For example:
142 define dumptype global {
144 record yes
145 index yes
146 }
147 define dumptype nocomp {
149 global # inherit the parameters in dumptype 'global'
150 compress none
151 }
152 Note that multiple inheritance is also supported by simply naming
154 multiple parent sections in a child. Parents are implicitly expanded in
155 place in a child, and the last occurrence of each parameter takes
156 precedence. For example,
157 define tapetype par1 {
159 comment "Parent 1"
160 filemark 8k
161 speed 300bps
162 length 200M
163 }
164 define tapetype par2 {
165 comment "Parent 2"
166 filemark 16k
167 speed 400bps
168 }
169 define tapetype child {
170 par1
171 par2
172 filemark 32k
173 }
174 In this example, 'child' will have a filemark of 32k, a speed of
175 400bps, and a length of 200M.
176
178 amrecover-changer string
179 Default: not set. Amrecover will use the changer if you use
180 'settape <string>' and that string is the same as the
181 amrecover-changer setting.
182 amrecover-check-label bool
184 Deprecated; amrecover always checks the label, and does not invoke
185 amrestore.
186 Default: on. Amrecover will call amrestore with the -l flag to
188 check the label.
189 amrecover-do-fsf bool
191 Deprecated; amrecover always uses fsf, and does not invoke
192 amrestore.
193 Default: on. Amrecover will call amrestore with the -f flag for
195 faster positioning of the tape.
196 autoflush no|yes|all
198 Default: no. Whether an amdump run will flush the dumps from
199 holding disk to tape. With yes, only dump matching the command line
200 argument are flushed. With all, all dump are flushed.
201 autolabel string [any] [other-config] [non-amanda] [volume-error]
203 [empty]
204 Default: not set. When set, this directive will cause Amanda to
205 automatically write an Amanda tape label to most volume she
206 encounters. This option is DANGEROUS because when set, Amanda may
207 erase near-failing tapes or tapes accidentally loaded in the wrong
208 slot.
209 When using this directive, specify the template for new tape
211 labels. The template can contains many variables that are
212 substituted by their values:
213 $c : config name
215 $o : org configuration
216 $b : barcode of the volume
217 $s : slot number, can specify a minimun number of digit:
218 $3s to get '001'
219 $m : meta label
220 $r : storage name
221 The template can contain some number of contiguous '%' characters,
223 which will be replaced with a generated number (000-999), or some
224 number of contiguous '!', which will be replaced with a generated
225 letter sequence (AAA-ZZZ). Be sure to specify enough '%' or '!'
226 characters that you do not run out of tape labels. Example:
227 "DailySet1-%%%", "DailySet1-!!!", "$c-%%%", "$m-%%%", "$m-$b"
228 The generared label can be used only if it match the labelstr
230 setting. The volume will not be used if the generated label doesn't
231 match the labelstr setting.
232 Note that many devices cannot distinguish an empty tape from an
234 error condition, so it may is often necessary to include
235 volume-error as an autolabel condition.
236 any
238 equivalent to 'other-config non-amanda volume-error empty'
239 other-config
241 Label volumes with a valid Amanda label that do not match our
242 labelstr. Danger: this may erase volumes from other Amanda
243 configurations without warning!
244 non-amanda
246 Label volumes which do not start with data that resembles an
247 Amanda header. Danger: this may erase volumes from other backup
248 applications without warning!
249 volume-error
251 Label volumes where an error occurs while trying to read the
252 label. Danger: this may erase arbitrary volumes due to
253 transient errors.
254 empty
256 Label volumes where a read returns 0 bytes.
257 bumpdays int
259 Default: 2 days. To insure redundancy in the dumps, Amanda keeps
260 filesystems at the same incremental level for at least bumpdays
261 days, even if the other bump threshold criteria are met.
262 The global setting of this parameter can be overwritten inside of a
264 dumptype-definition.
265 bumpmult float
267 Default: 1.5. The bump size multiplier. Amanda multiplies bumpsize
268 by this factor for each level. This prevents active filesystems
269 from bumping too much by making it harder to bump to the next
270 level. For example, with the default bumpsize and bumpmult set to
271 2.0, the bump threshold will be 10 Mbytes for level one, 20 Mbytes
272 for level two, 40 Mbytes for level three, and so on.
273 The global setting of this parameter can be overwritten inside of a
275 dumptype-definition.
276 bumppercent int
278 Default: 0. The minimum savings required to trigger an automatic
279 bump from one incremental level to the next, expressed as
280 percentage of the current size of the DLE (size of current level
281 0). If Amanda determines that the next higher backup level will be
282 this much smaller than the current level, it will do the next
283 level.
284 If this parameter is set to 0, the value of the parameter bumpsize
286 is used to trigger bumping.
287 The global setting of this parameter can be overwritten inside of a
289 dumptype-definition.
290 See also the options bumpsize, bumpmult and bumpdays.
292 bumpsize int
294 Default: 10 Mbytes. The minimum savings required to trigger an
295 automatic bump from one incremental level to the next, expressed as
296 size. If Amanda determines that the next higher backup level will
297 be this much smaller than the current level, it will do the next
298 level. The value of this parameter is used only if the parameter
299 bumppercent is set to 0.
300 The default unit is Kbytes if it is not specified.
302 The global setting of this parameter can be overwritten inside of a
304 dumptype-definition.
305 See also the options bumppercent, bumpmult and bumpdays.
307 changerdev string
309 Default: "dev/null". A tape changer configuration parameter. Usage
310 depends on the particular changer defined with the tpchanger
311 option.
312 changerfile string
314 Default: "changer". This option is deprecated; use the changerfile
315 in the changer section. Only chg-multi use it. A file where the
316 changer store its state.
317 columnspec string
319 default:
320 "HostName=0:-12:12,Disk=1:-11:11,Level=1:-1:1,OrigKB=1:-7:0,OutKB=1:-7:0,Compress=1:-6:1,DumpTime=1:-7:7,Dumprate=1:-6:1,TapeTime=1:-6:6,TapeRate=1:-6:1"
321 Defines the width of columns amreport should use. String is a
323 comma (',') separated list of triples. Each triple consists of four
324 parts which are separated by a equal sign ('=') and a colon (':')
325 (see the example). These four parts specify:
326 1. the name of the column, which may be:
328 Compress (compression ratio)
330 Disk (client disk name)
331 DumpRate (dump rate in KBytes/sec)
332 DumpTime (total dump time in hours:minutes)
333 HostName (client host name)
334 Level (dump level)
335 OrigKB (original image size in KBytes)
336 OutKB (output image size in KBytes)
337 TapeRate (tape writing rate in KBytes/sec)
338 TapeTime (total tape time in hours:minutes)
339 2. the amount of space to display before the column (used to get
341 whitespace between columns).
342 3. the width of the column itself. If set to a negative value, the
344 width will be calculated on demand to fit the largest entry in
345 this column.
346 4. the precision of the column, number of digit after the decimal
348 point for number.
349 Parts may be omitted, and will adopt a default value; trailing
351 colons may also be omitted.
352 Here is an example:
354 columnspec "Disk=1:18,HostName=0:10,OrigKB=::2,OutKB=1:7"
356 The above will display the disk information in 18 characters and
358 put one space before it. The hostname column will be 10 characters
359 wide with no space to the left. The Original KBytes print 2 decimal
360 digit. The output KBytes column is seven characters wide with one
361 space before it.
362 command-file string
364 Default: command_file. A file where amanda store information about
365 running job.
366 See amanda-command-file(5).
368 compress-index boolean
370 Default: yes. Compress all index files, this is useful to save
371 space in the indexdir but require more processing.
372 The compression ratio is generaly above 20x, it is faster to read
374 compressed index files because there is 20 times less data to read
375 from disk.
376 Changing this setting will uncompress/compress all index files.
378 connect-tries int
380 Default: 3. How many times the server will try a connection.
381 ctimeout int
383 Default: 30 seconds. Maximum amount of time that amcheck will wait
384 for each client host.
385 debug-auth int
387 Default: 0. Debug level of the auth module
388 debug-chunker int
390 Default: 0. Debug level of the chunker process
391 debug-days int
393 Default: 3. The number of days the debug files are kept.
394 debug-driver int
396 Default: 0. Debug level of the driver process
397 debug-dumper int
399 Default: 0. Debug level of the dumper process
400 debug-event int
402 Default: 0. Debug level of the event module
403 debug-holding int
405 Default: 0. Debug level of the holdingdisk module
406 debug-planner int
408 Default: 0. Debug level of the planner process
409 debug-protocol int
411 Default: 0. Debug level of the protocol module
412 debug-recovery int
414 Default: 1. Debug level of all recovery process
415 debug-taper int
417 Default: 0. Debug level of the taper process
418 device-output-buffer-size int
420 Default: 1280k. Controls the amount of memory used by Amanda to
421 hold data as it is read from the network or disk before it is
422 written to the output device. Higher values may be useful on fast
423 tape drives and optical media.
424 The default unit is bytes if it is not specified.
426 device-property string string
428 These options can set various device properties. See amanda-
429 devices(7) for more information on device properties and their
430 syntax. Both strings are always quoted; the first string contains
431 the name of the property to set, and the second contains its value.
432 For example, to set a fixed block size of 128k, write:
433 device-property "BLOCK_SIZE" "128k"
434 diskfile string
436 Default: "disklist". The file name for the disklist file holding
437 client hosts, disks and other client dumping information.
438 displayunit "k|m|g|t"
440 Default: "k". The unit used to print many numbers, k=kilo, m=mega,
441 g=giga, t=tera.
442 dtimeout int
444 Default: 1800 seconds. Amount of idle time per disk on a given
445 client that a dumper running from within amdump will wait before it
446 fails with a data timeout error.
447 dumpcycle int
449 Default: 10 days. The number of days in the backup cycle. Each disk
450 will get a full backup at least this often. Setting this to zero
451 tries to do a full backup each run.
452 Note
454 This parameter may also be set in a specific dumptype (see
455 below). This value sets the default for all dumptypes so must
456 appear in amanda.conf before any dumptypes are defined.
457 dumporder string
459 Default: "tttTTTTTTT". The priority order of each dumper:
460 s: smallest size
462 S: largest size
463 t: smallest time
464 T: largest time
465 b: smallest bandwidth
466 B: largest bandwidth
467 dumpuser string
469 Default: "amanda". The login name Amanda uses to run the backups.
470 The backup client hosts must allow access from the tape server host
471 as this user via .rhosts or .amandahosts, depending on how the
472 Amanda software was built.
473 eject-volume bool
475 Default: no. Set to yes if you want the volume to be ejected after
476 Amanda wrote data to it. It works only with some changer and
477 device.
478 etimeout int
480 Default: 300 seconds. Amount of time per estimate on a given client
481 that the planner step of amdump will wait to get the dump size
482 estimates (note: Amanda runs up to 3 estimates for each DLE). For
483 instance, with the default of 300 seconds and four DLE's, each
484 estimating level 0 and level 1 on client A, planner will wait up to
485 40 minutes for that machine. A negative value will be interpreted
486 as a total amount of time to wait per client instead of per disk.
487 flush-threshold-dumped int
489 Default: 0. Amanda will not begin writing data to a new volume
490 until the amount of data on the holding disk is at least this
491 percentage of the volume size and the criterion for
492 flush-threshold-scheduled is also met. In other words, Amanda will
493 not begin until the amount of data on the holding disk is greater
494 than the tape length times this parameter. This parameter may be
495 larger than 100%, for example to keep more recent dumps on the
496 holding disk for faster recovery.
497 Needless to say, your holding disk must be big enough that this
499 criterion could be satisfied. If the holding disk cannot be used
500 for a particular dump (because, for example, there is no remaining
501 holding space) then Amanda will disregard the constraint specified
502 by this setting and start a new volume anyway. Once writing to a
503 volume has begun, this constraint is not applied unless and until a
504 new volume is needed.
505 The value of this parameter may not exceed than that of the
507 flush-threshold-scheduled parameter.
508 flush-threshold-scheduled int
510 Default: 0. Amanda will not begin writing data to a new volume
511 until the sum of the amount of data on the holding disk and the
512 estimated amount of data remaining to be dumped during this run is
513 at least this percentage of the volume size and the criterion for
514 flush-threshold-dumped is also met. In other words, Amanda will not
515 begin until the inequality h + s > t × d is satisfied, where h is
516 the amount of data on the holding disk, s is the total amount of
517 data scheduled for this run but not dumped yet, t is the capacity
518 of a volume, and d is this parameter, expressed as a percentage.
519 This parameter may be larger than 100%.
520 Needless to say, your holding disk must be big enough that this
522 criterion could be satisfied. If the holding disk cannot be used
523 for a particular dump (because, for example, there is no remaining
524 holding space) then Amanda will disregard the constraint specified
525 by this setting and start a new volume anyway. Once writing to a
526 volume has begun, this constraint is not applied unless and until a
527 new volume is needed.
528 The value of this parameter may not be less than that of the
530 flush-threshold-dumped or taperflush parameters.
531 includefile string
533 Default: no default. The name of an Amanda configuration file to
534 include within the current file. Useful for sharing dumptypes,
535 tapetypes and interface definitions among several configurations.
536 Relative pathnames are relative to the configuration directory.
537 indexdir string
539 Default "/usr/adm/amanda/index". The directory where index files
540 (backup image catalogues) are stored. Index files are only
541 generated for filesystems whose dumptype has the index option
542 enabled.
543 infofile string
545 Default: "/usr/adm/amanda/curinfo". The file or directory name for
546 the historical information database. If Amanda was configured to
547 use DBM databases, this is the base file name for them. If it was
548 configured to use text formatted databases (the default), this is
549 the base directory and within here will be a directory per client,
550 then a directory per disk, then a text file of data.
551 inparallel int
553 Default: 10. The maximum number of backups that Amanda will attempt
554 to run in parallel. Amanda will stay within the constraints of
555 network bandwidth and holding disk space available, so it doesn't
556 hurt to set this number a bit high. Some contention can occur with
557 larger numbers of backups, but this effect is relatively small on
558 most systems.
559 interactivity string
561 Default: not set. The interactivity module Amanda should use to
562 interact with the user. See amanda-interactivity(7) for a list of
563 modules.
564 labelstr string | MATCH-AUTOLABEL
566 Default: MATCH-AUTOLABEL. The tape label constraint regular
567 expression. All tape labels generated (see amlabel(8)) and used by
568 this configuration must match the regular expression. All autolabel
569 variable can be used. The keywork MATCH-AUTOLABEL use the autolabel
570 template as expression.
571 If multiple configurations are run from the same tape server host,
573 it is helpful to set their labels to different strings (for
574 example, "DAILY[0-9][0-9]*" vs. "ARCHIVE[0-9][0-9]*") to avoid
575 overwriting each other's tapes.
576 label-new-tapes string
578 Deprecated, use autolabel option with options volume-error empty to
579 get equivalent behavior.
580 Default: not set. When set, this directive will cause Amanda to
582 automatically write an Amanda tape label to any blank tape she
583 encounters.
584 logdir string
586 Default: "/usr/adm/amanda". The directory for the amdump and log
587 files.
588 mailer string
590 Default found by configure. A mail program that can send mail with
591 'MAILER -s "subject" user < message_file'.
592 mailto string
594 Default: none. A space separated list of recipients for mail
595 reports. If not specified, amdump will not send any mail.
596 maxdumps int
598 Default: 1. The maximum number of backups from a single host that
599 Amanda will attempt to run in parallel. See also the inparallel
600 option.
601 Note that this parameter may also be set in a specific dumptype
603 (see below). This value sets the default for all dumptypes so must
604 appear in amanda.conf before any dumptypes are defined.
605 maxdumpsize int
607 Default: runtapes*tape-length. Maximum number of bytes the planner
608 will schedule for a run.
609 The default unit is Kbytes if it is not specified.
611 max-dle-by-volume int
613 Default: 1000000000. The maximum number of dle written to a single
614 volume.
615 meta-autolabel string
617 Default: not set. When set and if the changer support meta-label,
618 this directive will cause Amanda to automatically add a meta-label
619 to a meta-volume.
620 A meta-volume is a containers that contains many volumes, eg. a
622 removable hard-disk for use with chg-disk, each hard disk have many
623 slots (volume). The meta-label is the label to put on the
624 meta-volume.
625 When using this directive, specify the template for new meta
627 labels. The template can contains many variables that are
628 substituted by their values:
629 $c : config name
631 $o : org configuration
632 $r : storage name
633 The template can contain some number of contiguous '%' characters,
635 which will be replaced with a generated number (000-999), or some
636 number of contiguous '!', which will be replaced with a generated
637 letter sequence (AAA-ZZZ). Be sure to specify enough '%' characters
638 that you do not run out of meta labels. Example: "DailySet1-%%%",
639 "DailySet1-!!!", "$o-%%%", "$o-!!!", "$c-!!!",
640 netusage int
642 Default: 80000 Kbps. The maximum network bandwidth allocated to
643 Amanda, in Kbytes per second. See also the interface section.
644 org string
646 Default: "daily". A descriptive name for the configuration. This
647 string appears in the Subject line of mail reports. Each Amanda
648 configuration should have a different string to keep mail reports
649 distinct.
650 printer string
652 Printer to use when doing tape labels. See the lbl-templ tapetype
653 option.
654 property [append] string string+
656 These options can set various properties, they can be used by third
657 party software to store information in the configuration file. Both
658 strings are quoted; the first string contains the name of the
659 property to set, and the others contains its values. append
660 keyword append the values to the list of values for that property.
661 recovery-limit [ string | same-host | server]
663 Default: none (no limitations). This parameter limits the hosts
664 that may do recoveries. Hosts are identified by their authenticated
665 peer name, as described in amanda-auth(7); if this is not available
666 and the recovery-limit parameter is present, recovery will be
667 denied. The arguments to the parameter are strings giving host
668 match expressions (see amanda-match(7)) or the special keywords
669 same-host or server. The same-host keyword requires an exact match
670 to the hostname of the DLE being recovered. The server keyword
671 require the connection come from the fqdn of the server. Specifying
672 no arguments at all will disable all recoveries from any host.
673 Note that match expressions can be constructed to be forgiving of
675 e.g., fully-qualified vs. unqualified hostnames, but same-host
676 requires an exact match.
677 The error messages that appear in amrecover are intentionally vague
679 to avoid information leakage. Consult the amindexd debug log for
680 more details on the reasons a recovery was rejected.
681 Recovery limits can be refined on a per-DLE basis using the
683 dumptype parameter of the same name. Note that the default value
684 will apply to any dumpfiles for disks which no longer appear in the
685 disklist; thus leaving the global parameter at its default value
686 but setting it for all DLEs is not sufficient to maintain secure
687 backups.
688 report-format [append] string+
690 Default: Not set. The formats amdump, amflush and amvault use when
691 invoking amreport.
692 report-next-media boolean
694 Default: True if max-dle-by-volume is not set, False if it is set.
695 If the reporter must print the list of media expected for the next
696 run.
697 report-use-media boolean
699 Default: True if max-dle-by-volume is not set, False if it is set.
700 If the reporter must print the list of media used in the run.
701 req-tries int
703 Default: 3. How many times the server will resend a REQ packet if
704 it doesn't get the ACK packet.
705 reserve int
707 Default: 100. The part of holding-disk space that should be
708 reserved for incremental backups if no tape is available, expressed
709 as a percentage of the available holding-disk space (0-100). By
710 default, when there is no tape to write to, degraded mode
711 (incremental) backups will be performed to the holding disk. If
712 full backups should also be allowed in this case, the amount of
713 holding disk space reserved for incrementals should be lowered.
714 reserved-tcp-port int,int
716 Default: --with-low-tcpportrange or 512,1023. Reserved tcp port
717 that will be used (bsdtcp). Range is inclusive.
718 reserved-udp-port int,int
720 Default: --with-udpportrange or 512,1023. Reserved udp port that
721 will be used (bsd, bsdudp). Range is inclusive.
722 runspercycle int
724 Default: same as dumpcycle. The number of amdump runs in dumpcycle
725 days. A value of 0 means the same value as dumpcycle. A value of -1
726 means guess the number of runs from the tapelist(5) file, which is
727 the number of tapes used in the last dumpcycle days / runtapes.
728 runtapes int
730 Default: 1. The maximum number of tapes used in a single run. If a
731 tape changer is not configured, this option is not used and should
732 be commented out of the configuration file.
733 If a tape changer is configured, this may be set larger than one to
735 let Amanda write to more than one tape.
736 Note that this is an upper bound on the number of tapes, and Amanda
738 may use less.
739 send-amreport-on [ all | strange | error | never ]
741 Default: all. Specify which types of messages will trigger an email
742 from amreport. amreport is used by amdump and amflush.
743 all
745 Send an email on any message.
746 strange
748 Send an email on strange or error message. A strange message
749 occurs when the dump succeeded, but returned one or more errors
750 unknown to Amanda.
751 error
753 Send an email only on error messages.
754 never
756 Never send an email.
757 sort-index boolean
759 Default: no. Sort all index files, this make amrecover start faster
760 on big filesystem but it require more processing at backup time.
761 Changing this setting can sort all index files.
762 storage string+
764 Default: Same as the config name. The list of storages to use, the
765 dump will go to theses storages.
766 active-storage string+
768 Default: Same as the configured storage setting and the -ostorage=
769 command line setting. The list of storages where a dump can be put.
770 amdump keep the dump in holding disk if the storage is not also
771 listed in storage
772 tapebufs int
774 Default: 20. This option is deprecated; use the
775 device-output-buffer-size directive instead. tapebufs works the
776 same way, but the number specified is multiplied by the device
777 blocksize prior to use.
778 tapecycle int
780 Default: 15 tapes. Specifies the number of "active" volumes -
781 volumes that Amanda will not overwrite. While Amanda is always
782 willing to write to a new volume, it refuses to overwrite a volume
783 unless at least 'tapecycle -1' volumes have been written since.
784 It is considered good administrative practice to set the tapecycle
786 parameter slightly lower than the actual number of tapes in use.
787 This allows the administrator to more easily cope with damaged or
788 misplaced tapes or schedule adjustments that call for slight
789 adjustments in the rotation order.
790 Note: Amanda is commonly misconfigured with tapecycle equal to the
792 number of tapes per dumpcycle. In this misconfiguration, amanda may
793 erase a full dump before a new one is completed. Recovery is then
794 impossible. The tapecycle must be at least one tape larger than the
795 number of tapes per dumpcycle.
796 The number of tapes per dumpcycle is calculated by multiplying the
798 number of amdump runs per dump cycle runspercycle (the number of
799 amdump runs per dump cycle) and runtapes (the number of tapes used
800 per run). Typically tapecycle is set to two or four times the tapes
801 per dumpcycle.
802 tapedev string
804 Default: "null:". This parameter can either specify a device
805 (explicitly or by referencing a device definition - see amanda-
806 devices(7)) or a tape changer (explicitly or by referencing a
807 device definition - see amanda-changers(7)).
808 tapelist string
810 Default: "tapelist". The file name for the active tapelist(5).
811 Amanda maintains this file with information about the active set of
812 tapes.
813 taperalgo [ first | firstfit | largest | largestfit | smallest | last ]
815 Default: first. The algorithm used to choose which dump image to
816 send to the taper.
817 first
819 First in, first out.
820 firstfit
822 The first dump image that will fit on the current tape.
823 largest
825 The largest dump image.
826 largestfit
828 The largest dump image that will fit on the current tape.
829 smallest
831 The smallest dump image.
832 last
834 Last in, first out.
835 taperflush int
837 Default: 0. At the end of a run, Amanda will start a new tape to
838 flush remaining data if there is more data on the holding disk at
839 the end of a run than this setting allows; the amount is specified
840 as a percentage of the capacity of a single volume. In other words,
841 at the end of a run, Amanda will begin a new tape if the inequality
842 h > t × f is satisfied, where h is the amount of data remaining on
843 the holding disk from this or previous runs, t is the capacity of a
844 volume, and f is this parameter, expressed as a percentage. This
845 parameter may be greater than 100%.
846 The value of this parameter may not exceed that of the
848 flush-threshold-scheduled parameter.; autoflush must be set to
849 'yes' if taperflush is greater than 0.
850 taperscan string
852 Default: traditional. The taperscan module amanda should use to
853 find a tape to write to. See amanda-taperscan(7) for a list of
854 modules.
855 taper-parallel-write int
857 Default: 1. Amanda can write simultaneously up to that number of
858 volume at any given time. The changer must have as many drives.
859 tapetype string
861 Default: no default. The type of tape drive associated with tapedev
862 or tpchanger. This refers to one of the defined tapetypes in the
863 config file (see below), which specify various tape parameters,
864 like the length, filemark size, and speed of the tape media and
865 device.
866 tmpdir string
868 Default: none (system default). Set it to a directory with lots of
869 free space if sort in amindexd fail with 'No space left on device'.
870 tpchanger string
872 Default: not set. (deprecated) The tape changer to use. In most
873 cases, only one of tpchanger or tapedev is specified, although for
874 backward compatibility both may be specified if tpchanger gives the
875 name of an old changer script. See amanda-changers(7) for more
876 information on configuring changers.
877 unreserved-tcp-port int,int
879 Default: --with-tcpportrange or 1024,65535. Unreserved tcp port
880 that will be used (bsd, bsdudp). Range is inclusive.
881 usetimestamps bool
883 Default: Yes. Deprecated, the value is always Yes. This option
884 allows Amanda to track multiple runs per calendar day.
885 vault-storage string+
887 Default: not set. The list of storages to vault to.
888 After writing to the storages listed in the storage parameter,
890 amdump will automatically write all pending dumps to the vault
891 storage(s). (These dumps are queued for vaulting based on the vault
892 option specified in the definition section for the primary storage
893 and the dump-selection option specified on the vault storage.)
894 (amvault also uses the first storage in the vault-storage list as
896 its default destination storage.)
897
899 The amanda.conf file may define one or more holding disks used as
900 buffers to hold backup images before they are written to tape. The
901 syntax is:
902 define holdingdisk name {
903 holdingdisk-option holdingdisk-value
904 ...
905 }
906 The { must appear at the end of a line, and the } on its own line.
908 Name is a logical name for this holding disk.
910 The options and values are:
912 comment string
914 Default: not set. A comment string describing this holding disk.
915 chunksize int
917 Default: 1 Gb. Holding disk chunk size. Dumps larger than the
918 specified size will be stored in multiple holding disk files. The
919 size of each chunk will not exceed the specified value. However,
920 even though dump images are split in the holding disk, they are
921 concatenated as they are written to tape, so each dump image still
922 corresponds to a single continuous tape section.
923 The default unit is Kbytes if it is not specified.
925 If 0 is specified, Amanda will create holding disk chunks as large
927 as ((INT_MAX/1024)-64) Kbytes.
928 Each holding disk chunk includes a 32 Kbyte header, so the minimum
930 chunk size is 64 Kbytes (but that would be really silly).
931 Operating systems that are limited to a maximum file size of 2
933 Gbytes actually cannot handle files that large. They must be at
934 least one byte less than 2 Gbytes. Since Amanda works with 32 Kbyte
935 blocks, and to handle the final read at the end of the chunk, the
936 chunk size should be at least 64 Kbytes (2 * 32 Kbytes) smaller
937 than the maximum file size, e.g. 2047 Mbytes.
938 directory string
940 Default: "/dumps/amanda". The path to this holding area.
941 use int
943 Default: 0 Gb. Amount of space that can be used in this holding
944 disk area. If the value is zero, all available space on the file
945 system is used. If the value is negative, Amanda will use all
946 available space minus that value.
947 chunksize int
949 Default: 1 Gb. Holding disk chunk size. Dumps larger than the
950 specified size will be stored in multiple holding disk files. The
951 size of each chunk will not exceed the specified value. However,
952 even though dump images are split in the holding disk, they are
953 concatenated as they are written to tape, so each dump image still
954 corresponds to a single continuous tape section.
955 The default unit is Kbytes if it is not specified.
957 If 0 is specified, Amanda will create holding disk chunks as large
959 as ((INT_MAX/1024)-64) Kbytes.
960 Each holding disk chunk includes a 32 Kbyte header, so the minimum
962 chunk size is 64 Kbytes (but that would be really silly).
963 Operating systems that are limited to a maximum file size of 2
965 Gbytes actually cannot handle files that large. They must be at
966 least one byte less than 2 Gbytes. Since Amanda works with 32 Kbyte
967 blocks, and to handle the final read at the end of the chunk, the
968 chunk size should be at least 64 Kbytes (2 * 32 Kbytes) smaller
969 than the maximum file size, e.g. 2047 Mbytes.
970
972 The amanda.conf(5) file may define multiple sets of backup options and
973 refer to them by name from the disklist(5) file. For instance, one set
974 of options might be defined for file systems that can benefit from high
975 compression, another set that does not compress well, another set for
976 file systems that should always get a full backup and so on.
977 A set of backup options are entered in a dumptype section, which looks
979 like this:
980 define dumptype "name" {
981 dumptype-option dumptype-value
982 ...
983 }
984 The { must appear at the end of a line, and the } on its own line.
986 Name is the name of this set of backup options. It is referenced from
988 the disklist(5) file.
989 Some of the options in a dumptype section are the same as those in the
991 main part of amanda.conf(5). The main option value is used to set the
992 default for all dumptype sections. For instance, setting dumpcycle to
993 50 in the main part of the config file causes all following dumptype
994 sections to start with that value, but the value may be changed on a
995 section by section basis. Changes to variables in the main part of the
996 config file must be done before (earlier in the file) any dumptypes are
997 defined.
998 The dumptype options and values are:
1000 allow-split bool
1002 Default: true. If true, then dumps with this dumptype can be split
1003 on the storage media. If false, then the dump will be written in a
1004 single file on the media. See "Dump Splitting Configuration" below.
1005 application string
1007 No default. Must be the name of an application if program is set to
1008 APPLICATION. See APPLICATION SECTION below.
1009 auth string
1011 Default: "bsdtcp". Type of authorization to perform between tape
1012 server and backup client hosts. See amanda-auth(7) for more detail.
1013 amandad-path string
1015 Default: "$libexec/amandad". Specify the amandad path of the
1016 client, only use with rsh/ssh authentification.
1017 bumpdays int
1019 Default: 2 days. To insure redundancy in the dumps, Amanda keeps
1020 filesystems at the same incremental level for at least bumpdays
1021 days, even if the other bump threshold criteria are met.
1022 bumpmult float
1024 Default: 1.5. The bump size multiplier. Amanda multiplies bumpsize
1025 by this factor for each level. This prevents active filesystems
1026 from bumping too much by making it harder to bump to the next
1027 level. For example, with the default bumpsize and bumpmult set to
1028 2.0, the bump threshold will be 10 Mbytes for level one, 20 Mbytes
1029 for level two, 40 Mbytes for level three, and so on.
1030 bumppercent int
1032 Default: 0. The minimum savings required to trigger an automatic
1033 bump from one incremental level to the next, expressed as
1034 percentage of the current size of the DLE (size of current level
1035 0). If Amanda determines that the next higher backup level will be
1036 this much smaller than the current level, it will do the next
1037 level.
1038 If this parameter is set to 0, the value of the parameter bumpsize
1040 is used to trigger bumping.
1041 See also the options bumpsize, bumpmult and bumpdays.
1043 bumpsize int
1045 Default: 10 Mbytes. The minimum savings required to trigger an
1046 automatic bump from one incremental level to the next, expressed as
1047 size. If Amanda determines that the next higher backup level will
1048 be this much smaller than the current level, it will do the next
1049 level. The value of this parameter is used only if the parameter
1050 bumppercent is set to 0.
1051 The default unit is Kbytes if it is not specified.
1053 See also the options bumppercent, bumpmult and bumpdays.
1055 client-port [ int | string ]
1057 Default: "amanda". Specifies the port to connect to on the client.
1058 It can be a service name or a numeric port number.
1059 client-custom-compress string
1061 Default: none. The program to use to perform
1062 compression/decompression on the client; used with "compress client
1063 custom". Must not contain whitespace. Must accept -d to uncompress.
1064 client-decrypt-option string
1066 Default: -d. The option that can be passed to client-encrypt to
1067 make it decrypt instead. Must not contain whitespace.
1068 client-encrypt string
1070 Default: none. The program to use to perform encryption/decryption
1071 on the client; used with "encrypt client". Must not contain
1072 whitespace.
1073 client-username string
1075 Default: CLIENT_LOGIN. Specify the username to connect on the
1076 client, only use with rsh/ssh authentification.
1077 comment string
1079 Default: not set. A comment string describing this set of backup
1080 options.
1081 comprate float [, float ]
1083 Default: 0.50, 0.50. The expected full and incremental compression
1084 factor for dumps. It is only used if Amanda does not have any
1085 history information on compression rates for a filesystem, so
1086 should not usually need to be set. However, it may be useful for
1087 the first time a very large filesystem that compresses very little
1088 is backed up.
1089 compress [ none | client | server ] [ best | fast | custom ]
1091 Default: client fast. If Amanda does compression of the backup
1092 images, it can do so either on the backup client host before it
1093 crosses the network or on the tape server host as it goes from the
1094 network into the holding disk or to tape. Which place to do
1095 compression (if at all) depends on how well the dump image usually
1096 compresses, the speed and load on the client or server, network
1097 capacity, holding disk capacity, availability of tape hardware
1098 compression, etc.
1099 For either type of compression, Amanda also allows the selection of
1101 three styles of compression. best is the best compression
1102 available, often at the expense of CPU overhead. fast is often not
1103 as good a compression as best, but usually less CPU overhead. Or to
1104 specify custom to use your own compression method. (See dumptype
1105 custom-compress in example/amanda.conf for reference)
1106 So the compress options line may be one of:
1108 compress none
1110 compress client fast
1112 compress client best
1114 compress client custom
1116 Specify client-custom-compress "PROG"
1117 PROG must not contain white space and it must accept -d for
1119 uncompress.
1120 compress server fast
1122 compress server best
1124 compress server custom
1126 Specify server-custom-compress "PROG"
1127 PROG must not contain white space and it must accept -d for
1129 uncompress.
1130 Note that some tape devices do compression and this option has
1132 nothing to do with whether that is used. If hardware compression is
1133 used (usually via a particular tape device name or mt option),
1134 Amanda (software) compression should be disabled.
1135 dumpcycle int
1137 Default: 10 days. The number of days in the backup cycle. Each disk
1138 using this set of options will get a full backup at least this of
1139 ten. Setting this to zero tries to do a full backup each run.
1140 dump-limit [ server | same-host ]*
1142 Default: server. Specify which host can initiate a backup of the
1143 dle. With server, the server can initiate a backup with the amdump
1144 command. With same-host, the client can initiate a backup with the
1145 amdump_client command.
1146 encrypt [ none | client | server ]
1148 Default: not set. To encrypt backup images, it can do so either on
1149 the backup client host before it crosses the network or on the tape
1150 server host as it goes from the network into the holding disk or to
1151 tape.
1152 So the encrypt options line may be one of:
1154 encrypt none
1156 encrypt client
1158 Specify client-encrypt "PROG"
1159 PROG must not contain white space.
1161 Specify client-decrypt-option "decryption-parameter" Default:
1163 "-d"
1164 decryption-parameter must not contain white space.
1166 (See dumptype client-encrypt-nocomp in example/amanda.conf for
1168 reference)
1169 encrypt server
1171 Specify server-encrypt "PROG"
1172 PROG must not contain white space.
1174 Specify server-decrypt-option "decryption-parameter" Default:
1176 "-d"
1177 decryption-parameter must not contain white space.
1179 (See dumptype server-encrypt-fast in example/amanda.conf for
1181 reference)
1182 Note that current logic assumes compression then encryption during
1184 backup(thus decrypt then uncompress during restore). So specifying
1185 client-encryption AND server-compression is not supported. amcrypt
1186 which is a wrapper of aespipe is provided as a reference symmetric
1187 encryption program.
1188 estimate [ client | calcsize | server ]+
1190 Default: client. Determine the way Amanda estimates the size of
1191 each DLE before beginning a backup. This is a list of acceptable
1192 estimate methods, and Amanda applies the first method supported by
1193 the application. The methods are:
1194 client
1196 Use the same program as the dumping program. This is the most
1197 accurate method to do estimates, but it can take a long time.
1198 calcsize
1200 Use a faster program to do estimates, but the result is less
1201 accurate.
1202 server
1204 Use only statistics from the previous few runs to give an
1205 estimate. This very quick, but the result is not accurate if
1206 your disk usage changes from day to day. If this method is
1207 specified, but the server does not have enough data to make an
1208 estimate, then the option is internally moved to the end of the
1209 list, thereby preferring 'client' or 'calcsize' in this case.
1210 exclude [ list | file ][[optional][append][ string ]+]
1212 Default: file. Exclude is the opposite of include and specifies
1213 files that will be excluded from the backup. The format of the
1214 exclude expressions depends on the application, and some
1215 applications do not support excluding files at all.
1216 There are two exclude parameters, excludefile and excludelist.
1218 With excludefile, the string is an exclude expression. With
1219 excludelist , the string is a file name on the client containing
1220 GNU-tar exclude expressions. The path to the specified exclude list
1221 file, if present (see description of 'optional' below), must be
1222 readable by the Amanda user.
1223 All exclude expressions are concatenated in one file and passed to
1225 the application as an --exclude-from argument.
1226 For GNU-tar, exclude expressions must always be specified as
1228 relative to the top-level directory of the DLE, and must start with
1229 "./". See the manpages for individual applications for more
1230 information on supported exclude expressions.
1231 With the append keyword, the string is appended to the current
1233 list, without it, the string overwrites the list.
1234 If optional is specified for excludelist, then amcheck will not
1236 complain if the file doesn't exist or is not readable.
1237 For excludelist, if the file name is relative, the disk name being
1239 backed up is prepended. So if this is entered:
1240 exclude list ".amanda.excludes"
1241 the actual file used would be /var/.amanda.excludes for a backup of
1242 /var, /usr/local/.amanda.excludes for a backup of /usr/local, and
1243 so on.
1244 fallback-splitsize int
1246 Deprecated. See "Dump Splitting Configuration" below.
1247 Default: 10M. This specifies the part size used when no
1249 split-diskbuffer is specified, or when it is too small or does not
1250 exist, and thus the maximum amount of memory consumed for in-memory
1251 splitting. The default unit is Kbytes if it is not specified.
1252 holdingdisk [ never | auto | required ]
1254 Default: auto. Whether a holding disk should be used for these
1255 backups or whether they should go directly to tape. If the holding
1256 disk is a portion of another file system that Amanda is backing up,
1257 that file system should refer to a dumptype with holdingdisk set to
1258 never to avoid backing up the holding disk into itself.
1259 never|no|false|off
1261 Never use a holdingdisk, the dump will always go directly to
1262 tape. There will be no dump if you have a tape error.
1263 auto|yes|true|on
1265 Use the holding disk, unless there is a problem with the
1266 holding disk, the dump won't fit there or the medium doesn't
1267 require spooling (e.g., VFS device)
1268 required
1270 Always dump to holdingdisk, never directly to tape. There will
1271 be no dump if it doesn't fit on holdingdisk
1272 ignore boolean
1274 Default: no. Whether disks associated with this backup type should
1275 be backed up or not. This option is useful when the disklist file
1276 is shared among several configurations, some of which should not
1277 back up all the listed file systems.
1278 include [ list | file ][[optional][append][ string ]+]
1280 Default: file ".". There are two include lists, include file and
1281 include list. With include file , the string is a glob expression.
1282 With include list , the string is a file name on the client
1283 containing glob expressions.
1284 All include expressions are expanded by Amanda, concatenated in one
1286 file and passed to GNU-tar as a --files-from argument. They must
1287 start with "./" and contain no other "/".
1288 Include expressions must always be specified as relative to the
1290 head directory of the DLE.
1291 Note
1293 For globbing to work at all, even the limited single level, the
1294 top level directory of the DLE must be readable by the Amanda
1295 user.
1296 With the append keyword, the string is appended to the current
1297 list, without it, the string overwrites the list.
1298 If optional is specified for include list, then amcheck will not
1300 complain if the file doesn't exist or is not readable.
1301 For include list, If the file name is relative, the disk name being
1303 backed up is prepended.
1304 index boolean
1306 Default: no. Whether an index (catalogue) of the backup should be
1307 generated and saved in indexdir. These catalogues are used by the
1308 amrecover utility.
1309 kencrypt boolean
1311 Default: no. Whether the backup image should be encrypted by
1312 Kerberos as it is sent across the network from the backup client
1313 host to the tape server host.
1314 maxdumps int
1316 Default: 1. The maximum number of backups from a single host that
1317 Amanda will attempt to run in parallel. See also the main section
1318 parameter inparallel.
1319 maxpromoteday int
1321 Default: 10000. The maximum number of day for a promotion, set it 0
1322 if you don't want promotion, set it to 1 or 2 if your disks get
1323 overpromoted.
1324 max-warnings int
1326 Default: 20. The maximum number of error lines in the report for a
1327 dle. A value of '0' means unlimited. This is useful to reduce the
1328 size of the log file and the size of the report. All errors are put
1329 in separate files if a dle have more errors.
1330 priority [ low | medium | high ]
1332 Default: medium. When there is no tape to write to, Amanda will do
1333 incremental backups in priority order to the holding disk. The
1334 priority may be high (2), medium (1), low (0) or a number of your
1335 choice.
1336 program [ "DUMP" | "GNUTAR" | "APPLICATION" ]
1338 Default: "DUMP". The type of backup to perform. Valid values are:
1339 "DUMP"
1341 The native operating system backup program.
1342 "GNUTAR"
1344 To use GNU-tar or to do PC backups using Samba.
1345 "APPLICATION"
1347 To use an application, see the application option.
1348 property [append] [ hidden | visible ] string string+
1350 These options can set various properties, they can be used by third
1351 party software to store information in the configuration file. Both
1352 strings are quoted; the first string contains the name of the
1353 property to set, and the others contains its values. append
1354 keyword append the values to the list of values for that property.
1355 With hidden (the default), the property are not put in the amanda
1357 dump header and in the log/debug files. With visible, they are put
1358 in the amanda dump header and in the log/debug files. Use hidden if
1359 the property must be kept secret.
1360 record boolean
1362 Default: yes. Whether to ask the backup program to update its
1363 database (e.g. /etc/dumpdates for DUMP or
1364 /usr/local/var/amanda/gnutar-lists for GNUTAR) of time stamps. This
1365 is normally enabled for daily backups and turned off for periodic
1366 archival runs.
1367 recovery-limit [ server | same-host | string ]*
1369 Default: global value. This parameter overrides the global
1370 recovery-limit parameter for DLEs of this dumptype.
1371 retry-dump int
1373 Default: 2. The number of times a backup is tried in case of
1374 failure.
1375 script string
1377 No default. Must be the name of a script. You can have many script.
1378 See SCRIPT SECTION below.
1379 server-custom-compress string
1381 Default: none. The program to use to perform
1382 compression/decompression on the server; used with "compress server
1383 custom". Must not contain whitespace. Must accept -d to uncompress.
1384 server-decrypt-option string
1386 Default: -d. The option that can be passed to server-encrypt to
1387 make it decrypt instead. Must not contain whitespace.
1388 server-encrypt string
1390 Default: none. The program to use to perform encryption/decryption
1391 on the server; used with "encrypt server". Must not contain
1392 whitespace.
1393 skip-full boolean
1395 Default: no. If true and planner has scheduled a full backup, these
1396 disks will be skipped, and full backups should be run off-line on
1397 these days. It was reported that Amanda only schedules level 1
1398 incrementals in this configuration; this is probably a bug.
1399 skip-incr boolean
1401 Default: no. If true and planner has scheduled an incremental
1402 backup, these disks will be skipped.
1403 split-diskbuffer string
1405 Deprecated. See "Dump Splitting Configuration" below. Default: not
1407 set. When dumping a split dump in PORT-WRITE mode (usually meaning
1408 "no holding disk"), buffer the split chunks to a file in the
1409 directory specified by this option.
1410 ssh-keys string
1412 Default: not set. The key file the ssh auth will use, it must be
1413 the private key. If this parameter is not specified, then the
1414 default ssh key will be used.
1415 starttime int
1417 Default: not set. Backup of these disks will not start until after
1418 this time of day. The value should be hh*100+mm, e.g. 6:30PM
1419 (18:30) would be entered as 1830.
1420 strategy [ standard | nofull | noinc | skip | incronly ]
1422 Default: standard. Strategy to use when planning what level of
1423 backup to run next. Values are:
1424 standard
1426 The standard Amanda schedule.
1427 nofull
1429 Never do full backups, only level 1 incrementals.
1430 noinc
1432 Never do incremental backups, only full dumps.
1433 skip
1435 Treat this DLE as if it doesn't exist (useful to disable DLEs
1436 when sharing the disklist file between multiple
1437 configurations). Skipped DLEs will not be checked or dumped,
1438 and will not be matched by disklist expressions.
1439 incronly
1441 Only do incremental dumps. amadmin force should be used to
1442 tell Amanda that a full dump has been performed off-line, so
1443 that it resets to level 1.
1444 tag [append] string*
1446 Default: no default. Specify the tags that match the dump-selection
1447 of a storage.
1448 tape-splitsize int
1450 Deprecated. See "Dump Splitting Configuration" below.
1451 Default: not set. Split dump file on tape into pieces of a
1453 specified size. The default unit is Kbytes if it is not specified.
1454 The following dumptype entries are predefined by Amanda:
1456 define dumptype "no-compress" {
1457 compress none
1458 }
1459 define dumptype "compress-fast" {
1460 compress client fast
1461 }
1462 define dumptype "compress-best" {
1463 compress client best
1464 }
1465 define dumptype "srvcompress" {
1466 compress server fast
1467 }
1468 define dumptype "bsd-auth" {
1469 auth "bsd"
1470 }
1471 define dumptype "bsdtcp-auth" {
1472 auth "bsdtcp"
1473 }
1474 define dumptype "no-record" {
1475 record no
1476 }
1477 define dumptype "no-hold" {
1478 holdingdisk no
1479 }
1480 define dumptype "no-full" {
1481 skip-full yes
1482 }
1483 In addition to options in a dumptype section, one or more other
1485 dumptype names may be supplied as identifiers, which make this dumptype
1486 inherit options from other previously defined dumptypes. For instance,
1487 two sections might be the same except for the record option:
1488 define dumptype "normal" {
1489 comment "Normal backup, no compression, do indexing"
1490 no-compress
1491 index yes
1492 maxdumps 2
1493 }
1494 define dumptype "testing" {
1495 comment "Test backup, no compression, do indexing, no recording"
1496 "normal"
1497 record no
1498 }
1499 Amanda provides a dumptype named global in the sample amanda.conf file
1501 that all dumptypes should reference. This provides an easy place to
1502 make changes that will affect every dumptype, although you must be
1503 careful that every dumptype explicitly inherits from the global
1504 dumptype - Amanda does not do so automatically.
1505
1507 The amanda.conf file may define multiple types of tape media and
1508 devices. The information is entered in a tapetype section, which looks
1509 like this in the config file:
1510 define tapetype "name" {
1511 tapetype-option tapetype-value
1512 ...
1513 }
1514 The { must appear at the end of a line, and the } on its own line.
1516 Name is the name of this type of tape medium/device. It is referenced
1518 from the tapetype option in the main part of the config file.
1519 The tapetype options and values are:
1521 blocksize int
1523 Default: 32 kbytes. How much data will be written in each tape
1524 record, expressed in kbytes. This is similar to the BLOCK_SIZE
1525 device property, but if the blocksize is not a multiple of 1024
1526 bytes, then this parameter cannot be used to specify it, and the
1527 property must be used instead.
1528 comment string
1530 Default: not set. A comment string describing this set of tape
1531 information.
1532 filemark int
1534 Default: 1 kbytes. How large a file mark (tape mark) is, measured
1535 in kbytes. If the size is only known in some linear measurement
1536 (e.g. inches), convert it to kbytes using the device density.
1537 lbl-templ string
1539 Default: not set. A PostScript template file used by amreport to
1540 generate labels. Several sample files are provided with the Amanda
1541 sources in the example directory. See the amreport(8) man page for
1542 more information.
1543 length int
1545 Default: 2000 kbytes. How much data will fit on a tape, expressed
1546 in kbytes.
1547 Note that this value is only used by Amanda to schedule which
1549 backups will be run. Once the backups start, Amanda will continue
1550 to write to a tape until it gets an error, regardless of what value
1551 is entered for length (but see amanda-devices(7) for exceptions).
1552 part-cache-dir string
1554 Default: none. The directory in which part-cache files can be
1555 written when caching on disk. See "Dump Splitting Configuration"
1556 below.
1557 part-cache-max-size int
1559 Default: none. The maximum part size to use when caching is in
1560 effect. This is used to limit the part size when disk or memory
1561 space for caching is constrained. This value must be greater than
1562 zero.
1563 part-cache-type [ none | disk | memory ]
1565 Default: none. When part caching is required, this parameter
1566 specifies the type of caching that will be used. The options
1567 include no caching (none), in which case a failed part will cause
1568 the entire dump to fail; on-disk caching (disk), for which
1569 part-cache-dir must be set properly; and in-memory caching
1570 (memory), which on most systems severely restrains the size of the
1571 part that can be written. See "Dump Splitting Configuration" below.
1572 part-size int
1574 If this is set to zero (default), then no splitting will take
1575 place, and the entire dump will fail, if end-of-medium is
1576 encountered before the dump is complete, unless the device property
1577 LEOM is true, and the device can detect EOM. See "Dump Splitting
1578 Configuration" below.
1579 readblocksize int
1581 Default: 32 kytes How much data will be read in each tape record.
1582 This can be used to override a device's block size for reads only.
1583 This may be useful, for example, in reading a tape written with a
1584 256k block size when Amanda is configured to use 128k blocks. This
1585 unusual feature is not supported by all operating systems and tape
1586 devices.
1587 The default unit is Kbytes if it is not specified.
1589 speed int
1591 Default: 200 bps. How fast the drive will accept data, in bytes per
1592 second. This parameter is NOT currently used by Amanda.
1593 In addition to options, another tapetype name may be supplied as an
1595 identifier, which makes this tapetype inherit options from another
1596 tapetype. For instance, the only difference between a DLT4000 tape
1597 drive using Compact-III tapes and one using Compact-IV tapes is the
1598 length of the tape. So they could be entered as:
1599 define tapetype "DLT4000-III" {
1600 comment "DLT4000 tape drives with Compact-III tapes"
1601 length 12500 mbytes # 10 Gig tapes with some compression
1602 filemark 2000 kbytes
1603 speed 1536 kps
1604 }
1605 define tapetype "DLT4000-IV" {
1606 "DLT4000-III"
1607 comment "DLT4000 tape drives with Compact-IV tapes"
1608 length 25000 mbytes # 20 Gig tapes with some compression
1609 }
1610
1612 The amanda.conf file may define multiple types of network interfaces.
1613 The information is entered in an interface section, which looks like
1614 this:
1615 define interface "name" {
1616 interface-option interface-value
1617 ...
1618 }
1619 The { must appear at the end of a line, and the } on its own line.
1621 name is the name of this type of network interface. It is referenced
1623 from the disklist file.
1624 If a src-ip is specified, then the connection should be from an
1626 interface with that IP. The system decide which interface to use if
1627 src-ip is not specified. You can add route at the system level to do
1628 more specific routing.
1629 The section do not impose limits on the bandwidth that will actually be
1631 taken up by Amanda. Amanda computes the estimated bandwidth each file
1632 system backup will take based on the estimated size and time, then
1633 compares that plus any other running backups with the limit as another
1634 of the criteria when deciding whether to start the backup. Once a
1635 backup starts, Amanda will use as much of the network as it can leaving
1636 throttling up to the operating system and network hardware.
1637 The interface options and values are:
1639 comment string
1641 Default: not set. A comment string describing this set of network
1642 information.
1643 src-ip string
1645 The IP address to use when sending a request to an amanda client.
1646 use int
1648 Default: 80000 Kbps. The speed of the interface in Kbytes per
1649 second.
1650 In addition to options, another interface name may be supplied as an
1652 identifier, which makes this interface inherit options from another
1653 interface. At the moment, this is of little use.
1654
1656 The amanda.conf file may define multiple types of application. The
1657 information is entered in a application section, which looks like this:
1658 define application "name" {
1659 application-option application-value
1660 ...
1661 }
1662 The { must appear at the end of a line, and the } on its own line.
1664 name is the name of this type of application. It is referenced from the
1666 dumptype
1667 The application options and values are:
1669 client-name string
1671 No default, specifies an application name that is in the
1672 amanda-client.conf on the client. The setting from that application
1673 will be merged with the current application. If client-name is set
1674 then it is an error if that application is not defined on the
1675 client.
1676 If client-name is not set then the merge is done with the
1678 application that have the name equal to the plugin. eg. if the
1679 plugin is 'amgtar', then the setting from the application 'amgtar'
1680 is used if it is defined.
1681 comment string
1683 Default: not set. A comment string describing this application.
1684 plugin string
1686 No default. Must be set to the name of the program. This program
1687 must be in the $libexecdir/amanda/application directory on the
1688 client.
1689 property [append] [priority] string string+
1691 No default. You can set property for the application, each
1692 application have a different set of property. Both strings are
1693 quoted; the first string contains the name of the property to set,
1694 and the others contains its values. append keyword append the
1695 values to the list of values for that property. priority keyword
1696 disallow the setting of that property on the client.
1697
1699 The amanda.conf file may define multiple types of script. The
1700 information is entered in a script section, which looks like this:
1701 define script "name" {
1702 script-option script-value
1703 ...
1704 }
1705 The { must appear at the end of a line, and the } on its own line.
1707 name is the name of this type of script. It is referenced from the
1709 dumptype
1710 The script options and values are:
1712 client-name string
1714 No default, specifies a script name that is in the
1715 amanda-client.conf on the client. The setting from that script will
1716 be merged with the currect script. If client-name is set then it is
1717 an error if that script is not defined on the client.
1718 If client-name is not set then the merge is done with the script
1720 that have the name equal to the plugin. eg. if the plugin is
1721 'amlog-script', then the setting from the script 'amlog-script' is
1722 used.
1723 comment string
1725 Default: not set. A comment string describing this script.
1726 execute-on execute_on [,execute_on]*
1728 No default. When the script must be executed, you can specify many
1729 of them:
1730 pre-amcheck
1732 Execute before the amcheck command for all dle. Can only be run
1733 on server.
1734 pre-dle-amcheck
1736 Execute before the amcheck command for the dle.
1737 pre-host-amcheck
1739 Execute before the amcheck command for all dle for the client.
1740 post-amcheck
1742 Execute after the amcheck command for all dle. Can only be run
1743 on server.
1744 post-dle-amcheck
1746 Execute after the amcheck command for the dle.
1747 post-host-amcheck
1749 Execute after the amcheck command for all dle for the client.
1750 pre-estimate
1752 Execute before the estimate command for all dle. Can only be
1753 run on server.
1754 pre-dle-estimate
1756 Execute before the estimate command for the dle.
1757 pre-host-estimate
1759 Execute before the estimate command for all dle for the client.
1760 post-estimate
1762 Execute after the estimate command for all dle. Can only be run
1763 on server.
1764 post-dle-estimate
1766 Execute after the estimate command for the dle.
1767 post-host-estimate
1769 Execute after the estimate command for all dle for the client.
1770 pre-backup
1772 Execute before the backup command for all dle. Can only be run
1773 on server.
1774 pre-dle-backup
1776 Execute before the backup command for the dle.
1777 pre-host-backup
1779 Execute before the backup command for all dle for the client.
1780 It can't be run on client, it must be run on server
1781 post-backup
1783 Execute after the backup command for all dle. Can only be run
1784 on server.
1785 post-dle-backup
1787 Execute after the backup command for the dle.
1788 post-host-backup
1790 Execute after the backup command for all dle for the client. It
1791 can't be run on client, it must be run on server
1792 pre-recover
1794 Execute before any level is recovered.
1795 post-recover
1797 Execute after all levels are recovered.
1798 pre-level-recover
1800 Execute before each level recovery.
1801 post-level-recover
1803 Execute after each level recovery.
1804 inter-level-recover
1806 Execute between two levels of recovery.
1807 If you recover level 0 and 2 of the disk /usr with amrecover, it
1809 will execute:
1810 script --pre-recover
1811 script --pre-level-recover --level 0
1812 #recovering level 0
1813 script --post-level-recover --level 0
1814 script --inter-level-recover --level 0 --level 2
1815 script --pre-level-recover --level 2
1816 #recovering level 2
1817 script --post-level-recover --level 2
1818 script --post-recover
1819 execute-where [ client | server ]
1821 Default: client. Where the script must be executed, on the client
1822 or server.
1823 order int
1825 Default: 5000. Scripts are executed in that order, it is useful if
1826 you have many scripts and they must be executed in a spefific
1827 order.
1828 plugin string
1830 No default. Must be set to the name of the program. This program
1831 must be in the $libexecdir/amanda/application directory on the
1832 client and/or server.
1833 property [append] [priority] string string+
1835 No default. You can set property for the script, each script have a
1836 different set of property. Both strings are quoted; the first
1837 string contains the name of the property to set, and the others
1838 contains its values. append keyword append the values to the list
1839 of values for that property. priority keyword disallow the setting
1840 of that property on the client.
1841 single-execution boolean
1843 Default: no. The script is executed for each dle. If yes, the
1844 script is executed one time only.
1845
1847 Backend storage devices are specified in amanda.conf in the form of
1848 "device" sections, which look like this:
1849 define device name {
1850 comment "comment (optional)"
1851 tapedev "device-specifier"
1852 device-property "prop-name" "prop-value"
1853 ...
1854 }
1855 The { must appear at the end of a line, and the } on its own line.
1857 name is the user-specified name of this device. It is referenced from
1859 the global tapedev parameter. The device-specifier specifies the device
1860 name to use; see amanda-devices(7). As with most sections, the comment
1861 parmeter is optional and only for the user's convenience.
1862 An arbitrary number of device-property parameters can be specified.
1864 Again, see amanda-devices(7) for information on device properties.
1865
1867 Changers are described in amanda.conf in the form of "changer"
1868 sections, which look like this:
1869 define changer name {
1870 comment "comment (optional)"
1871 tpchanger "changer-spec"
1872 changerdev "device-name"
1873 changerfile "state-file"
1874 ...
1875 }
1876 The { must appear at the end of a line, and the } on its own line.
1878 name is the user-specified name of this device. The remaining
1880 parameters are specific to the changer type selected.
1881 The tpchanger and changerfile can use '$t' to substitute the name of
1883 the changer.
1884 See amanda-changers(7) for more information on configuring changers.
1886
1888 The amanda.conf file may define multiple interactivyt methods, although
1889 only one will be used - that specified by the interactivity parameter.
1890 The information is entered in a interactivity section, which looks like
1891 this:
1892 define interactivity name {
1893 interactivity-option interactivity-value
1894 ...
1895 }
1896 The { must appear at the end of a line, and the } on its own line.
1898 name is the user-specified name of this interactivity. The remaining
1900 parameters are specific to the interactivity type selected.
1901 The interactivity options and values are:
1903 comment string
1905 Default: not set. A comment string describing this interactivity.
1906 plugin string
1908 No default. Must be set to the name of the interactivity module, as
1909 described in amanda-interactivity(7).
1910 property [append] string string+
1912 No default. You can set arbitrary properties for the interactivity.
1913 Each interactivity module has a different set of properties. The
1914 first string contains the name of the property to set, and the
1915 others contains its values. All strings should be quoted. The
1916 append keyword appends the given values to an existing list of
1917 values for that property.
1918 See amanda-interactivity(7) for more information on configuring
1920 interactivity methods.
1921
1923 The amanda.conf file may define multiple taperscan methods, it is set
1924 with the global taperscan parameter or in the storage section. The
1925 information is entered in a taperscan section, which looks like this:
1926 define taperscan name {
1927 taperscan-option taperscan-value
1928 ...
1929 }
1930 The { must appear at the end of a line, and the } on its own line.
1932 name is the user-specified name of this taperscan. The remaining
1934 parameters are specific to the taperscan type selected.
1935 The taperscan options and values are:
1937 comment string
1939 Default: not set. A comment string describing this taperscan.
1940 plugin string
1942 No default. Must be set to the name of the taperscan module. See
1943 amanda-taperscan(7) for a list of defined taperscan modules.
1944 property [append] string string+
1946 No default. Operates just like properties for interactivity
1947 methods, above.
1948 See amanda-taperscan(7) for more information on configuring taperscan.
1950
1952 The amanda.conf file may define multiple policy, it is set with the
1953 policy parameter of the storage section. A policy name CONFIG_NAME is
1954 automaticaly created. The information is entered in a policy section,
1955 which looks like this:
1956 define policy name {
1957 policy-option policy-value
1958 ...
1959 }
1960 The { must appear at the end of a line, and the } on its own line.
1962 name is the user-specified name of this policy. The remaining
1964 parameters are specific to the policy type selected.
1965 The policy options and values are:
1967 comment string
1969 Default: not set. A comment string describing this policy.
1970 retention-days int
1972 Default: 0. A volume is kept for that number of days before it can
1973 be reused.
1974 retention-full int
1976 Default: 0. A volume is kept if it contains a FULL not older then
1977 retention-full days.
1978 retention-recover int
1980 Default: 0. Keep all volumes needed to recover all files up to
1981 retention-recover days ago, if it was a level 2, also keep previous
1982 level 1 and level 0.
1983 retention-tapes int
1985 Default: global tapecycle-1. The latest used retention-tapes volume
1986 can't be reused. You must have more than retention-tapes volumes
1987 before a volume can be re-used.
1988
1990 The amanda.conf file may define multiple storage, the default storages
1991 are set with the storage parameter. The vaulting storage are set with
1992 the the vault-storage parameter. A storage name CONFIG_NAME is created
1993 if the global storage is not set. The information is entered in a
1994 storage section, which looks like this:
1995 define storage name {
1996 storage-option storage-value
1997 ...
1998 }
1999 The { must appear at the end of a line, and the } on its own line.
2001 name is the user-specified name of this storage. The remaining
2003 parameters are specific to the storage type selected.
2004 The storage options and values are:
2006 autoflush no|yes|all
2008 Default: value of the global autoflush.
2009 autolabel string [any] [other-config] [non-amanda] [volume-error]
2011 [empty]
2012 Default: value of the global autolabel.
2013 comment string
2015 Default: not set. A comment string describing this storage.
2016 device-output-buffer-size int
2018 Default: value of the global device-output-buffer-size.
2019 dump-selection [string | ALL] [ ALL | FULL | INCR ]
2021 Default: no default. The dump-selection specify which dump will be
2022 written to the storage. The first field is the tag, either it is
2023 ALL and all DLEs matches or it is a tag string and a dle match only
2024 if it have that tag. The second field is the level, it can be ALL
2025 for all level, FULL for level 0 only or INCR for level > 0 only.
2026 eject-volume bool
2028 Default: value of the global eject-volume.
2029 erase-on-failure bool
2031 Default: NO. Automatically erase a volume if nothing useful was
2032 written to it. This is useful to reuse the volume sooner.
2033 erase-on-full bool
2035 Default: NO. Automatically erase a no-retention volume if the vtape
2036 area become full.
2037 erase-volume bool
2039 Default: NO. Automatically erase the volume when the policy expire.
2040 This is useful to free space on vtape or s3 devices or to allow
2041 another storage to use that volume.
2042 flush-threshold-dumped int
2044 Default: value of the global flush-threshold-dumped.
2045 flush-threshold-scheduled int
2047 Default: value of the global flush-threshold-scheduled.
2048 interactivity string
2050 Default: value of the global interactivity.
2051 labelstr string
2053 Default: value of the global labelstr.
2054 max-dle-by-volume int
2056 Default: value of the global max-dle-by-volume.
2057 meta-autolabel string
2059 Default: value of the global meta-autolabel.
2060 policy string
2062 Default: CONFIG_NAME. Define the policy to use.
2063 report-next-media boolean
2065 Default: value of the global report-next-media.
2066 report-use-media boolean
2068 Default: value of the global report-use-media.
2069 runtapes int
2071 Default: value of the global runtapes.
2072 set-no-reuse bool
2074 Default: no. If set to yes, a volume is marked as no-reuse after it
2075 is written.
2076 tapedev string
2078 Default: value of the global tapedev.
2079 This parameter can either specify a device (explicitly or by
2081 referencing a device definition - see amanda-devices(7)) or a tape
2082 changer (explicitly or by referencing a device definition - see
2083 amanda-changers(7)).
2084 tapepool string
2086 Default: CONFIG_NAME. Some characters are substituted:
2087 $o : org configuration
2089 $c : config name
2090 $r : storage name
2091 taperalgo [ first | firstfit | largest | largestfit | smallest | last ]
2093 Default: value of the global taperalgo.
2094 taperflush int
2096 Default: value of the global taperflush.
2097 taperscan string
2099 Default: value of the global taperscan.
2100 taper-parallel-write int
2102 Default: value of the global taper-parallel-write.
2103 tapetype string
2105 Default: value of the global tapetype.
2106 tpchanger string
2108 Default: value of the global tpchanger.
2109 vault storage int
2111 Will vault all dumps from this storage to the new storage X days
2112 after the dumps. You can have multiple vault entry.
2113
2115 Amanda can "split" dumps into parts while writing them to storage
2116 media. This allows Amanda to recover gracefully from a failure while
2117 writing a part to a volume, by simply selecting a new volume and
2118 re-writing the dump from the beginning of the failed part. Parts also
2119 allow Amanda to seek directly to the required data, although this
2120 functionality is not yet used.
2121 In order to support re-writing from the beginning of a failed part,
2123 Amanda must have access to the contents of the part after it has been
2124 partially written. If the dump is being read from holding disk, then
2125 the part contents are available there. Otherwise, the part must be
2126 cached, and this can be done memory or on disk. In either of the latter
2127 cases, the cache must have enough space to hold an entire part.
2128 Because it is common for a single Amanda configuration to use both
2130 holding-disk (FILE-WRITE) and direct (known as PORT-WRITE) dumps,
2131 Amanda allows the configuration of different split sizes for the two
2132 cases. This allows, for example, for a part size appropriate to large
2133 tapes when performing FILE-WRITE dumps, with a part size limited by
2134 available disk or memory when performing PORT-WRITE dumps.
2135 Selecting a proper split size is a delicate matter. If the parts are
2137 too large, substantial storage space may be wasted in failed parts. If
2138 too small, large dumps will be split into innumerable tiny dumpfiles,
2139 adding to restoration complexity; furthermore, an excess of filemarks
2140 will cause slower tape drive operation and reduce the usable space on
2141 tape. A good rule of thumb is 1/10 of the size of a volume of storage
2142 media.
2143 In versions of Amanda through 3.1.*, splitting was controlled by the
2145 dumptype parameters tape-splitsize, split-diskbuffer, and
2146 fallback-splitsize. These keywords had confusing and non-intuitive
2147 interactions, and have since been deprecated.
2148 If the deprecated keywords are not present, subsequent versions of
2150 Amanda use the dumptype parameter allow-split to control whether a DLE
2151 can be split, and the tapetype parameters part-size, part-cache-type,
2152 part-cache-dir, and part-cache-max-size. The part-size specifies the
2153 "normal" part size, while the part-cache-* parameters describe how to
2154 behave when caching is required (on PORT-WRITE). Full details on these
2155 parameters are given above.
2156
2158 amanda(8), amanda-applications(7), amanda-auth(7), amanda-changers(7),
2159 amanda-client.conf(5), amanda-command-file(5), amanda-devices(7),
2160 amanda-interactivity(7), amanda-scripts(7), amanda-taperscan(7),
2161 amgetconf(8), amadmin(8)
2162 The Amanda Wiki: : http://wiki.zmanda.com/
2164
2166 James da Silva <jds@amanda.org>
2167 Stefan G. Weichinger <sgw@amanda.org>
2169Amanda 3.5.1 12/01/2017 AMANDA.CONF(5)